The brightness changes induced in red and green tests by an achromatic surround were studied using a nulling method. The amplitude of the nulling modulation was found to increase approximately linearly with the temporal contrast of an inducing field of constant average luminance. The amplitude of the nulling modulation relative to the amplitude of the inducing modulation increased with increasing test luminance and decreased with increasing mean luminance of the surround. For a given contrast, the amplitude of the nulling modulation increased with the amplitude of the inducing modulation. Nevertheless, the amplitude of the nulling modulation was comparatively larger when the mean luminance of the surround was close to the test mean luminance. These results were fit by the model developed by Spehar et al. [Vision Res. 36 (1996)]. In this model, the total brightness induced by a surround is given by a weighted spatial summation of the induced effects from each point in the surround with local luminance gain controls in the test and surround fields. It is assumed that the magnitude of induction from each surround element is gain controlled bq the difference between the mean luminance of the test and the individual surround elements. The values of the nulling modulation were practically the same for the green and fire rad test, On the countrary, with a heterochromatic brightness matching method, the luminance changes needed to match the test brightness depended on test chromaticity. At equal surround luminance, rite largest changes in luminance were for the green test and the smallest were for the blue test, the red test being at an intermediate level. The influence of the test chromaticity in the direct heterochromatic brightness matching experiment shows a contribution of the chromatic channel. The modulation nulling conceals this contribution, because the task performed by the observer seems to involve the achromatic channel only.